genomematt / xenomapper Goto Github PK

hisat and other spliced aligners use the XS tag to identify the strand that the match occurs on. Need to implement logic that allows identification of multimapping reads in this context.

sam header is written to correct file, but not sequences

Comment field in SAM header is not tab separated from comment resulting in invalid sam files

While running:
xenomappability --mapped_test_data xenomapper/GRCh38_100_reads_GRCh38_specific.sam > GRCh38_100_reads_GRCh38_specific.wig

Got following error:

Traceback (most recent call last):
File "/home/fernandr/biotools/bin/xenomappability", line 11, in
sys.exit(main())
File "/home/fernandr/biotools/lib/python3.4/site-packages/xenomapper/mappability.py", line 302, in main
single_end_mappability_from_sam(samfile=args.mapped_test_data)
File "/home/fernandr/biotools/lib/python3.4/site-packages/xenomapper/mappability.py", line 183, in single_end_mappability_from_sam
name_pos = int(name.split('')[1])
ValueError: invalid literal for int() with base 10: 'KI270721v1'_

I guess it's related to chromosome names in reference. As suggested, I used:

GCA_000001405.15_GRCh38_no_alt_analysis_set.fna

xenomappability is a tool for creating mapability wiggle files that reflect the paired end and multi species nature of the final number more accurately than the commonly used single end mapability tracks.

Error produced by chromosome/sequence names containing underscores when running xenommappability --mapped_test_data

Reads with forward state of 'unresolved' where the AS == XS are incorrectly assigned in output files

Hello,

I have successfully ran xenomapper on xenograft PE WGS data aligned using bwa mem hs37d5 (primary) and mm10 (secondary):

xenomapper --primary_bam "$PRIMARY".merged.dedup.rnsort.test.bam     --secondary_bam "$SECONDARY".merged.dedup.rnsort.test.bam     --primary_specific >(samtools view -bS - > "$PRIMARY"-specific.bam)     --secondary_specific >(samtools view -bS - > "$SECONDARY"-specific.bam)     --unassigned >(samtools view -bS - > "$DIR$SAMPLE"-unassigned.bam)
--------------------------------------------------------------------------------
Read Count Category Summary

|       Category                                     |     Count       |
|:--------------------------------------------------:|:---------------:|
|  primary_multi                                     |             29  |
|  primary_specific                                  |            377  |
|  secondary_multi                                   |              3  |
|  secondary_specific                                |             23  |
|  unresolved                                        |              8  |

However, when I use the --paired option (on the same data) I receive the following error:

xenomapper --primary_bam "$PRIMARY".merged.dedup.rnsort.test.bam     --secondary_bam "$SECONDARY".merged.dedup.rnsort.test.bam     --primary_specific >(samtools view -bS - > "$PRIMARY"-specific.bam)     --secondary_specific >(samtools view -bS - > "$SECONDARY"-specific.bam)     --unassigned >(samtools view -bS - > "$DIR$SAMPLE"-unassigned.bam)     --paired
Traceback (most recent call last):
  File "/home/561/lo6532/.local/bin/xenomapper", line 8, in <module>
    sys.exit(main())
  File "/home/561/lo6532/.local/lib/python3.7/site-packages/xenomapper/xenomapper.py", line 729, in main
    tag_func=tag_func)
  File "/home/561/lo6532/.local/lib/python3.7/site-packages/xenomapper/xenomapper.py", line 398, in main_paired_end
    for line1,line2 in readpairs:
  File "/home/561/lo6532/.local/lib/python3.7/site-packages/xenomapper/xenomapper.py", line 80, in getBamReadPairs
    assert line1[0] == line2[0]
AssertionError

I should also mention, after running xenomapper the primary_specific bam file is returning only single reads which seems strange . Are you able to comment on this?

Any help would be greatly appreciated :) Thanks!

Modifications to deal with ALT hits flagged with 0x800 (supplementary) and a variable number of mappings for the same read id.
Xenomapper needs to collect all these as a group, use the maximum score to compare to mouse, and allocate them to the correct read type as a block.
Tests should include cases of unique on alt and multimapping on alt

currently version id changes require a change in consistency test hash

"Corrisponding" should be "corresponding" in README.md and xenomapper.py.

regards,
Paul Harrison

Python 3.7 changes behaviour of iterators that reach end of file. Requires enhanced EOF state handling